[libcxx][pretty printers] Report not being able to trace test program

[DavidSpickett]

Summary

If you don't have ptrace permissions this test will fail to run
silently, this adds a check for that and anything else that
might do similar things.

The output will now be:

FAILED test program did not run correctly, check gdb warnings

/usr/bin/gdb: warning: Couldn't determine a path for the index cache
directory.
No symbol table is loaded.  Use the "file" command.
warning: Error disabling address space randomization: Operation not
permitted
warning: Could not trace the inferior process.
warning: ptrace: Operation not permitted

error: command failed with exit status: 255

We already have a feature to check for a compatible python enabled
gdb, so I think it's reasonable to check for this at test runtime.

Note that this is different to the catch all at the end of the test
script. That would be a case where you can trace but something else
made it stop mid way that wasn't our test breakpoints.

[DavidSpickett]

I found this testing with the CI docker image. By default my docker doesn't give ptrace permissions to the container, --cap-add=SYS_PTRACE fixes that.

If you think having that capability for the existing CI machines is going to be an issue I can come up with a way to feature test this up front instead.

None of them get as far as running this test so it's hard to tell whether they already have it. I was manually building a focal based image that's how I got it to run.

libcxx/test/libcxx/gdb/gdb_pretty_printer_test.py

 """
 
 from __future__ import print_function
 import re
 import gdb
 import sys
 
 test_failures = 0
+# Sometimes the inital run command can fail to trace the process.
+# (e.g. you don't have ptrace permissions)
+# In these cases gdb still sends us an exited event so we cannot
+# see what "run" printed to check for a warning message, since
+# we get taken to our exit handler before we can look.
+# Instead check that at least one test has been run by the time
+# we exit.
+has_run_tests = False
 
 
 class CheckResult(gdb.Command):
 
     def __init__(self):
         super(CheckResult, self).__init__(
             "print_and_compare", gdb.COMMAND_DATA)
 
     def invoke(self, arg, from_tty):
         try:
+            has_run_tests = True
+
             # Stack frame is:
             # 0. StopForDebugger
             # 1. ComparePrettyPrintToChars or ComparePrettyPrintToRegex
             # 2. TestCase
             compare_frame = gdb.newest_frame().older()
             testcase_frame = compare_frame.older()
             test_loc = testcase_frame.find_sal()
             # Use interactive commands in the correct context to get the pretty
def _get_value_string(self, compare_frame, testcase_frame):
             s = gdb.execute("p " + clean_expression_str, to_string=True)
         if sys.version_info.major == 2:
             return s.decode("utf-8")
         return s
 
 
 def exit_handler(event=None):
     global test_failures
-    if test_failures:
+    global has_run_tests
+
+    if not has_run_tests:
+        print("FAILED test program did not run correctly, check gdb warnings")
+        test_failures = -1
+    elif test_failures:
         print("FAILED %d cases" % test_failures)
     exit(test_failures)
 
 
 # Start code executed at load time
 
 # Disable terminal paging
 gdb.execute("set height 0")